**8. Experimental results and conclusions**

This section summarises the clinical implications and shows how the obtained experimental findings in the previous works [83, 84, 97] and their significance have led to developing explanatory AI models. For example **Table 3** illustrated the promising results obtained by the proposed Stepwise approach discussed in [33, 84].

In **Table 4**, the prediction performance of the underlying patterns of complications for these patients within the discovered subgroup dataset (was introduced in [83] as DS1 and discovered using the descriptive strategy) was analysed and compared to all patients belonged to DS (the raw T2DM dataset). It also suggested that DS1 (by personalising patients) could be considered as a dataset with less uncertainty than DS. In order to describe the inference problem in this chapter, the causal relationships seemed to be a reliable option to represent static and dynamic correlations between T2DM risk factors. The causal inference has a greater focus on distinguishing causes from other associations than on uncovering detailed temporal relationships. Therefore, in this work ([83]), several predictive strategies in order to


#### **Table 3.**

and life-threatening comorbidities [64]. Fowler and co-authors in [7] researched type 2 Diabetic American patients. This research utilised T2DM key risk factors such as HbA1c, SBP, and DBP to investigate relationships among complications such as HYP, NEP, RET, and NEU. In addition, LIV is a severe phenotype of diabetes and associated with T2DM complications, especially NEU [92]. Litwak analysed Russian diabetic patients in [93] which referred to the influence of macrovascular and micro-vascular disease on one anther. For example, important features in T2DM dataset such as blood pressure, HDL, lipid, BMI, and HbA1c influence diabetic patients' complications. They also revealed that HDL has a negative effect on HYP, NEP, NEU, and RET, whereas HbA1c negatively associated with HYP. Again, a study conducted by Ramachandran [94] referred to the high prevalence of NEU and RET in Type 2 diabetes in India. Similarly, research in [76] suggested that most of the diabetic patients have objective evidence for some variety of NEU, but only a few of them have identified by symptoms. This research also showed that there is a strong association among NEP, NEU, and RET. This study only concentrates on five binary complications as the predictive target classes in a binary classification problem (with two categories of classes: "high" or "low" risk). Furthermore, a complication class value of low risk (zero) represents a patient visit in which the complication is not present; otherwise, it is at high risk (one). For instance, a complication class value of zero represents a patient visit in which the complication is not present; otherwise, it is one. Alternatively, other risk factors associated with a patient (symptoms/clinical tests) are abstracted in the multi-class

*The description of the T2DM clinical features, risk factors, control values, and the discretised states [83].*

**Node ID Target complication Diagnosis outcome<sup>a</sup> Clinical risk class<sup>b</sup>**

*The description of T2DM target complication, clinical node control values, and discretised states [83].*

**Node ID T2DM risk factors Control value<sup>a</sup> Discretised value<sup>b</sup>** HbA1c (HBA) 6.6 � 1.2 (%) {low,medium,high} Body Mass Index (BMI) 26.4 � 2.4 kg*=*m<sup>2</sup> ð Þ {low,medium,high} Creatinine (CRT) 0.9 � 0.2 mg ð Þ *=*dL {low,medium,high} Cholesterol (COL) 0.9 � 0.2 mg ð Þ *=*dL {low,medium,high} High-Density Lipoprotein (HDL) 1.1 � 0.3 mmol ð Þ *=*l {low,medium,high} Diastolic Blood Pressure (DBP) 91 � 12 mmHg ð Þ {low,medium,high} Systolic Blood Pressure (SBP) 148 � 19 mmHg ð Þ {low,medium,high} Smoking Habit (SMK) {0,1,2} {low,medium,high}

 Retinopathy (RET) {Negative,Positive} {low,high} Neuropathy (NEU) {Negative,Positive} {low,high} Nephropathy (NEP) {Negative,Positive} {low,high} Liver Disease (LIV) {Negative,Positive} {low,high} Hypertension (HYP) {Negative,Positive} {low,high}

*Type 2 Diabetes - From Pathophysiology to Cyber Systems*

*a*

*b*

*a*

**212**

**Table 2.**

*(Mean* � *SD). <sup>b</sup> low, medium, high.*

**Table 1.**

*Negative test result, Positive test result.*

*Low clinical risk, High clinical risk.*

*Comparison of our new and enhanced stepwise IC\*LS approach in [97] with its previous version (stepwise IC\*) in [33] and without latent variable in [32].*


**Table 4.**

*The overall prediction accuracy of T2DM complications for patients in DS is compared to DS1.*

test whether the descriptive approaches have contributed to improving the prediction performance of the ordering patterns of complications.

non-stationary, complex, and incomplete data. For instance, the pre-processing approaches, statistical analysis, temporal phenotype, MCI algorithm and the DBNs model could be applied to another complex data (e.g., COVID-19). Therefore, in a new project, a similar patient model to this research was mainly employed, which primarily concentrated on helping healthcare staff in their understanding of how

*Predicting Type 2 Diabetes Complications and Personalising Patient Using Artificial…*

University City London (UCL) associated with the BHF Alan Turing Institute jointly funded research project with the collaborators of the project in UCL and GSK. This project aims to develop a computational tool to investigate the action of drug compounds for the treatment of cardiovascular disease and type 2 diabetes which involves: firstly, the construction of a cardiovascular disease (CVD) and Type-II diabetes (T2D) relevant metabolic measures networks, using repeated measures. Secondly, the combination of different causal networks on the same set of metabolic measures. Lastly, the integration to the system of available drug targets

In the current work as a Post-Doctorate research fellow at Brunel University and

I thank the following individuals for their expertise and assistance throughout all aspects of this study and for their insightful suggestions and careful reading of the manuscript. Dr. Stephen Swift, Dr. Mahir Arzoky and clinicians/scientist at Pavia hospital; especially Lucia Saachi and Luca Chiovato for providing the dataset and

Life Science Department, College of Health, Medicine and Life Sciences, Brunel

© 2020 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

COVID-19 spread and how they could be better prepared.

*DOI: http://dx.doi.org/10.5772/intechopen.94228*

and disease information for testing CVD and T2D drugs.

**Acknowledgements**

their professional advice.

**Author details**

**215**

Leila Yousefi\* and Allan Tucker

University London, United Kingdom

provided the original work is properly cited.

\*Address all correspondence to: leila.yousefi@brunel.ac.uk
